Yahoo!’s new service, Pipes, is generally lauded as something very original. And it’s true: although at first it seems similar to an RSS mixer (check those out in my RSS mixer roundup), it’s a completely different breed of product.

First, you are able to combine feeds (and other data) in a structured, algorithmic, hierarchical way. Example. You can take two news sources, and combine them into one. Then do the same thing with another three news sources. Then you can filter the two resulting feeds for a keyword, and then combine it with something else. See the possibilities?

Secondly, it’s visual. You can see the structure of your data as you would see connections within a database. The data is shown in the form of interconnected windows spread on a grid. This is best illustrated with a screenshot.

Thirdly, it’s complex. Casual users probably won’t be able to use many of its features. It looks more like a programmer’s tool than a service for end users. It’s powerful, but power comes with a price.

And lastly, it’s versatile. It can take all sorts of data - RSS feeds, OPML, various searches, text and numerical input, URL input, data from Flickr…the possibilities are endless. It also has many operators like truncate, union, sort, “for each” loop. Finally, you can take the results and make them public. Other users can clone your pipe and use it as they wish.

In a nutshell, Pipes is an advanced tool which you can use to build structured data from various sources, do stuff with it and use the product in any way you want. You can make a website which will aggregate data from the web and present it in the exact way you want it to. In the past, to build a mashup you need to be a programmer; now all you have to do is get familiar with Pipes. It’s a kind of service that won’t make a boom right away, but it will make a big impact. I can imagine that in a couple of months hundreds of new services based on Pipes will show up. Great new service, and a truly great move from Yahoo!.